Endometrial hyperplasia, abnormal, nonmalignant, proliferation of the uterine endometrium, is a common condition associated with the development of endometrial cancer and is classified as simple hyperplasia, complex hyperplasia and hyperplasia with atypia. The latter two are the most severe forms and are commonly treated with progestin therapy or hysterectomy. A proportion of hyperplasias do not respond to progestin treatment, but there are no known clinical predictors to identify this subgroup. In addition, precise estimates of persistence or progression of disease to guide treatment decisions are lacking. The proposed cohort study consists of female members of a large prepaid health care organization in Western Washington, ages 30-80, who developed complex hyperplasia or hyperplasia with atypia (n-1,600) during 1993-2003 who did not receive a hysterectomy at the time of diagnosis. The primary aims are to asses the following in women with a) complex hyperplasia and b) hyperplasia with atypia: 1) the likelihood of histologic regression, persistence, or progression of the condition and 2) the rate of subsequent hysterectomy or endometrial cancer. Women who have received treatment with a progestin will be compared for the above outcomes to women not treated with a progestin. The secondary aim is to: explore the association of immunohistochemical markers as predictors of regression of disease. These specific aims will be accomplished by using automated pathology, outpatient visit, and enrollment data to identify a cohort of women with all types of hyperplasia. Among women with complex hyperplasia, and hyperplasia with atypia, we will: a) verify pathologic diagnoses by standardized pathology review; b) link automated pharmacy, surgical and medical databases to the pathology information; c) identify treatment -progestin, hysterectomy or expectant management; d) review medical records to collect information about potential confounding factors, including race, height, weight, age of menopause, prior hormone use and cancer diagnoses; and e) conduct a nested case-control study (n-400) to evaluate the association of immunohistochemical markers with disease persistence/progression.